This research paper establishes a rigorous taxonomy for analyzing prompt attacks and defenses in Large Language Models, introducing JAILBREAKDB and a standar...
Level: advanced
By Hanbin Hong and 8 other authors
Category: discussion